What can electron paramagnetic resonance tell us about the Si/SiO2 system?

Abstract

Electron paramagnetic resonance (EPR) measurements of Si/SiO2 systems began over 30 years ago. Most EPR studies of Si/SiO2 systems have dealt with two families of defects: Pb centers and E′ centers. Several variants from each group have been observed in a wide range of Si/SiO2 samples. Some of the most basic aspects of this extensive, body of work remain controversial. EPR is an extraordinary powerful analytical tool quite widely utilized in chemistry, biomedical research, and solid state physics. Although uniquely well suited for metal–oxide–silicon (MOS) device studies, its capabilities are not widely understood in the MOS research and development community. The impact of EPR has been limited in the MOS community by a failure of EPR spectroscopists to effectively communicate with other engineers and scientists in the MOS community. In this article we hope to, first of all, ameliorate the communications problem by providing a brief but quantitative introduction to those aspects of EPR which are most relevant to MOS systems. We review, critically, those aspects of the MOS/EPR literature which are most relevant to MOS technology and show how this information can be used to develop physically based reliability models. Finally, we briefly review EPR work dealing with impurity defects in oxide thin films.